2-Keto-L-gulonic acid ("2-KLG") is a significant intermediate in the preparation of L-ascorbic acid (vitamin C), an essential nutrient. 2-KLG has been synthesized in the past on an industrial scale using the Reichstein method (Helvetica Chimica Acta 17:311 (1934)). This method, however, has a number of disadvantages for commercial application, including the use of large quantities of solvents and the involvement of a number of complex reaction steps.
Accordingly, as an alternative to the Reichstein method, a number of processes employing one or more microorganisms have been developed to produce 2-KLG by fermentation. U.S. Pat. No. 2,421,611, for example, discloses a method involving microbial oxidation of D-glucose to 5-keto-D-gluconic acid, followed by chemical or microbial reduction to L-idonic acid and subsequent microbial oxidation to 2-KLG. Japanese Patent Publication Nos. 39-14493, 53-25033, 56-15877 and 59-35290, for example, disclose similar processes involving the microbial oxidation of D-glucose to 2,5-diketo-D-gluconic acid followed by microbial or chemical reduction to 2-KLG.
These methods, however, also suffer from a number of disadvantages that reduce their usefulness in commercial production of 2-KLG. For example, the chemical reduction steps in these methods (i.e. the reduction of 5-keto-D-gluconic acid to L-idonic acid and 2,5-diketo-D-gluconic acid to 2-KLG) are accompanied by problems with controlling the stereochemistry of reduction (thus producing D-gluconic acid and 2-keto-D-gluconic acid, respectively, as byproducts) which, in turn, reduces the yield of 2-KLG. Alternatively, when this reduction is performed by one or more microorganisms, excess sugar is required to provide an energy source for the reduction, which also reduces the yield of 2-KLG.
In view of these problems, an alternate pathway has been employed for the fermentative production of 2-KLG, which involves only oxidation of L-sorbose to 2-KLG via a sorbosone intermediate. A number of processes have been developed using this pathway that employ a wide range of microorganisms from the genera Gluconobacter, such as Gluconobacter oxydans (U.S. Pat. Nos. 4,935,359; 4,960,695; 5,312,741; and 5,541,108), Pseudogluconobacter, such as Pseudogluconobacter saccharoketogenes (U.S. Pat. Nos. 4,877,735; European Patent No. 221 707), Pseudomonas, such as Pseudomonas sorbosoxidans (U.S. Pat. Nos. 4,933,289 and 4,892,823), and mixtures of microorganisms from these and other genera, such as Acetobacter, Bacillus, Serratia, Mycobacterium, and Streptomyces (U.S. Pat. Nos. 3,912,592; 3,907,639; and 3,234,105).
These processes, however, suffer from certain disadvantages that limit their usefulness for commercial production of 2-KLG. For example, the processes referenced above that employ G. oxydans also require the presence of an additional "helper" microbial strain, such as Bacillus megaterium, or commercially unattractive quantities of yeast or growth components derived from yeast in order to produce sufficiently high levels of 2-KLG for commercial use. Similarly, the processes that employ Pseudogluconobacter can require medium supplemented with expensive and unusual rare earth salts or the presence of a helper strain, such as B. megaterium, and/or the presence of yeast in order to achieve commercially suitable 2-KLG concentrations and efficient use of sorbose substrate. Other processes that employ Pseudomonas sorbosoxidans also include commercially unattractive quatities of yeast or yeast extract in the medium.
Accordingly, there is a need in the art for microorganism strains which efficiently produce 2-KLG, but without many of the problems associated with the state of the art.